• The current COVID-19 pandemia, caused by the SARS-CoV-2 virus, has caused the death of more than 3.2 million people worldwide.

• Studies in convalescent patients has shown a decline of antibody-mediated response a few months after infection.

• CD8+ T cells are critical for efficient clearance of virus-infected cells and long-term protection against SARS-CoV-2.

• Saini et al have analyzed T cell recognition towards SARS-CoV-2-derived peptides in COVID-19 patients and healthy individuals using DNA-barcoded pMHC multimers combined with a T cell phenotyping.

• In this project, we analyze the immunophenotypes of CD8 T cells that can recognize peptides from the SARS-CoV-2 genome
• We compare COVID-19 patients with healthy donors (HD) never exposed to the virus.
• Furthermore, we compare the phenotypes to the severity of the disease.
• We also compare the phenotype of the SARS-CoV-2 specific CD8 T cells with T cells that recognize common viruses.

Fig. 1-Experimental pipeline for T cell identification from COVID-19 patients.

Phenotype data

• Phenotyping of immune cells provides imformation abouts the activation status of the cells
• This is obtained through antibody staining and flow cytometry, which transforms a fluorescent signal into a voltage.
• Cell populations of specific phenotype is gated in FlowJo software, and frequencies of positive cells can afterwards exported as an .xlsx file

Tidying the data set

• Change NA values
• Add cohort group as a new variable
• Collapse multiple columns into two columns: Gating and Fraction
• Splitting the gating path into two new columns: Parent and Last population

## # A tibble: 1,886 x 8
##    SampleID  cohort_type Hospital_status Gating          T_cell Parent_populati…
##    <chr>     <chr>       <chr>           <chr>           <chr>  <chr>           
##  1 Patient_1 Patient     Hospitalized    lymphocytes/si… CD8    CD8+            
##  2 Patient_1 Patient     Hospitalized    lymphocytes/si… CD8    SARS_multimer+  
##  3 Patient_1 Patient     Hospitalized    lymphocytes/si… CD8    SARS_multimer+  
##  4 Patient_1 Patient     Hospitalized    lymphocytes/si… CD8    SARS_multimer+  
##  5 Patient_1 Patient     Hospitalized    lymphocytes/si… CD8    SARS_multimer+  
##  6 Patient_1 Patient     Hospitalized    lymphocytes/si… CD8    SARS_multimer+  
##  7 Patient_1 Patient     Hospitalized    lymphocytes/si… CD8    SARS_multimer+  
##  8 Patient_1 Patient     Hospitalized    lymphocytes/si… CD8    SARS_multimer+  
##  9 Patient_1 Patient     Hospitalized    lymphocytes/si… CD8    SARS_multimer+  
## 10 Patient_1 Patient     Hospitalized    lymphocytes/si… CD8    SARS_multimer+  
## # … with 1,876 more rows, and 2 more variables: Last_population <chr>,
## #   Fraction <dbl>

• FCS files are the raw data files obtained with flow cytometry.
• In addition to manual gating in FlowJo, these files can be used for clustering and visualizing the cells’ expression levels of multiple parameters in 2 dimensions with UMAP (Uniform Manifold Approximation and Projection).

Pipeline

• Read multiple FCS files (flowCORE) and downsample
• Tidying the data set by renaming the samples and variables
• Scale transformation
• Calculate UMAP coordinates (uwot)

• Wrangle data for visualization
• Generate plot using ggplot - Boxplot, dotplot, hexplot (2d density plot)
• Multiple plots generated for different surface markers
• Statistical test was added to the visualization

• list the main results!!